Why nobody knows who will win the 2024 election
Almost two weeks to Election Day, I feel more uncertain about this year’s result than any election I’ve covered professionally. Some of that is due to the polls – they’re really tight – but it’s also because for every good signal for Donald Trump, there seems to be a good one for Kamala Harris.
Many Americans believe this election has a lot on the line. And yet to me, it’s still a race with multiple potential outcomes – from a clear Harris victory to a contest that can’t be projected until late into election night (or week) to a decisive Trump win.
Let’s start with a simple proposition: Harris’ easiest path victory runs through winning the “blue wall” states of Michigan, Pennsylvania and Wisconsin. If she loses in the Sun Belt battlegrounds (Arizona, Georgia, Nevada and North Carolina), carrying the three Great Lakes states, along with Nebraska’s 2nd District and all the other states Joe Biden won in 2020, would get Harris to exactly 270 electoral votes.
Polling averages in Michigan, Pennsylvania and Wisconsin show the margins between Harris and Trump at under a point right now. Were that to hold until Election Day, it would be the first time in at least 50 years that any of those three states had final average margins within a point.
More than that, the margins in those states have been consistently close since Harris got into the race in July. Neither Trump nor Harris has ever led in any of those states by 5 points or more, which mirrors the national polling. It’s the first time in over 60 years that no candidate has led by 5 points or more nationally at any point in the race.
Many Republicans are hoping that the close polling points to a blowout win for Trump next month. The former president outperformed the polls significantly in 2016 and 2020. If Trump did so again, he would likely coast to victory, getting over 300 electoral votes.
I would be cautious, however, about assuming that a polling miss would benefit Trump. Going back to 1972, we’ve never had three presidential cycles in a row in which the same party benefited from a state polling miss. In fact, surveys in the key battlegrounds in 2022 underestimated Democrats. If we had a polling miss like two years ago, Harris would probably win over 300 electoral votes.
Battleground state polling averages are far from perfect. The average miss is 3.4 points since 1972, and 5% of the time they miss by more than 9.4 points. Even an average miss in the key battlegrounds could turn the election into a blowout.
You might be tempted to guess the direction of a potential polling miss by the macro trends. Biden’s approval rating is awful. No incumbent’s party has ever won another term when the president’s approval rating is this low. No incumbent’s party has ever won another term when so few people think the country is heading in the right direction.
But Trump may be exactly the wrong candidate to take advantage of these structural advantages. If he wins, he’d be the second least-liked candidate to do so since pollsters started tracking candidate popularity in the mid- 20th century. The only presidential winner who was less popular was Trump himself, in 2016. Remember also that Republicans underperformed in the 2022 midterms, even while many of the macro indicators pointed in their direction.
Take party registration. The trends aren’t as clear as they may appear at first glance. Republicans have been gaining on Democrats in all the key battleground states over the past four years. That, along with national party identification trends, would usually translate to Republicans cruising to victory this year.
But it’s not clear how many registered Republicans will cast ballots for Trump. It’s possible, as the most recent New York Times/Siena College poll of Pennsylvania shows, that Harris will win a higher share of Democrats than Trump will of Republicans. So with registered Democrats outnumbering registered Republicans in the Keystone State, such an outcome would mean that Harris very likely wins Pennsylvania.
What makes it all the more interesting is that we have a close race despite a lot of Americans changing their voting patterns from four years ago. Trump looks destined to put up one of the best performances for a Republican presidential nominee with Black voters in many years. This is especially true among young Black men.
Yet, Harris seems to be doing better among White women than any Democratic presidential nominee this century. While her gains are not as great as Trump’s among Black voters, White women make up a much larger portion of the electorate. Therefore, these shifts may cancel each other out to a large degree.
This means the election is likely to come down to the few voters who remain undecided.
More than two-thirds of likely voters believe that this is the most important election of their lifetime, including 72% of Trump supporters and 70% of Harris’ backers. The 5% of voters who are currently undecided will determine which of them walks away from this election happy.
Ironically, only 24% of those undecided voters agree that this election is the most important of their lifetime.
How painful it must be for those who really care about the election that it will be decided by many who don’t.
For more CNN news and newsletters create an account at CNN.com